Question answering using entity references in unstructured data
Methods, systems, and computer-readable media are provided for collective reconciliation. In some implementations, a query is received, wherein the query is associated at least in part with a type of entity. One or more search results are generated based at least in part on the query. Previously generated data is retrieved associated with at least one search result of the one or more of search results, the data comprising one or more entity references in the at least one search result corresponding to the type of entity. The one or more entity references are ranked, and an entity result is selected from the one or more entity references based at least in part on the ranking. An answer to the query is provided based at least in part on the entity result.
1. A method performed by one or more processors comprising:
obtaining, for a query, search results for the query and a ranking of the search results,
wherein the query is a natural language query,
wherein the search results are generated based on the query, and
wherein the ranking of the search results is based on relevance of the search results to the query;
identifying, based on at least one term of the query, that the query is associated with a particular type of entity, of a plurality of candidate types of entities,
wherein the particular type of entity defines a particular categorization that encompasses multiple entities, and
wherein the particular type of entity is one of a person type, a location type, or a date type;
selecting one or more top ranked search results, as determined based on the ranking of the search results;
selecting an entity reference from content of at least one of the selected one or more of the top ranked search results,
wherein the entity reference includes particular text that refers to a particular entity, and
wherein selecting the entity reference is responsive to the entity reference being included in the content of the selected one or more of the top ranked search results and is responsive to the entity reference being of the particular type of entity identified based on the at least one term of the query; and
providing the entity reference for display along with, but visually distinguished from, at least the top ranked search results, wherein providing the entity reference for display at a computing device is responsive to a submission of the query from the computing device, and wherein providing the entity reference visually distinguished from at least the top ranked search results is responsive to selecting the entity reference.
2. The method of claim 1 , wherein providing the entity reference for display along with, but visually distinguished from, at least the top ranked search results comprises providing the entity reference positioned above the top ranked search results.
3. The method of claim 1 , wherein the entity reference is provided for display as part of a natural language answer to the query, and further comprising:
generating the natural language answer to the query based on the entity reference and based on the query, wherein generating the natural language answer to the query based on the entity reference and based on the query comprises including at least one query term, from the query, in the natural language answer and including the particular text, of the entity reference, in the natural language answer.
4. The method of claim 1 , further comprising:
identifying an alternate entity reference from the content of at least one of the one or more of the top ranked search results,
wherein the alternate entity reference includes alternate particular text that refers to an alternate particular entity that is also of the particular type of entity, and
wherein selecting the entity reference comprises selecting the entity reference over the alternate entity reference.
5. The method of claim 4 , wherein selecting the entity reference over the alternate entity reference comprises:
determining a first ranking signal for the entity reference and a second ranking signal for the alternate entity reference;
ranking the entity reference and the alternate entity reference based on the first ranking signal and the second ranking signal; and
selecting the entity reference based on the ranking.
6. The method of claim 5 , wherein determining the first ranking signal for the entity references comprises determining the first ranking signal based on frequency of occurrence of the entity reference in the selected one or more of the top ranked search results.
7. The method of claim 1 , wherein the particular type of entity is the person type.
8. A system comprising:
memory storing instructions; and
one or more processors executing the instructions to perform operations comprising:
obtaining, for a query, search results for the query and a ranking of the search results,
wherein the query is a natural language query,
wherein the search results are generated based on the query, and
wherein the ranking of the search results is based on relevance of the search results to the query;
identifying, based on at least one term of the query, that the query is associated with a particular type of entity, of a plurality of candidate types of entities,
wherein the particular type of entity defines a particular categorization that encompasses multiple entities, and
wherein the particular type of entity is one of a person type,
a location type, or a date type;
selecting a top ranked search result, as determined based on the ranking of the search results;
selecting an entity reference from content of the selected top ranked search result,
wherein the entity reference includes particular text that refers to a particular entity, and
wherein selecting the entity reference is responsive to the entity reference being included in the content of the selected top ranked search result and is responsive to the entity reference being of the particular type of entity identified based on the at least one term of the query; and
providing the entity reference for display along with, but visually distinguished from, the top ranked search result, wherein providing the entity reference for display at a computing device is responsive to a submission of the query from the computing device, and wherein providing the entity reference visually distinguished from at least the top ranked search results is responsive to selecting the entity reference.
9. The system of claim 8 , wherein providing the entity reference for display along with, but visually distinguished from, the top ranked search result comprises providing the entity reference positioned above the top ranked search result.
10. The system of claim 8 , wherein the entity reference is provided for display as part of a natural language answer to the query, and the operations performed by the one or more processors in executing the instructions further comprise:
generating the natural language answer to the query based on the entity reference and based on the query, wherein generating the natural language answer to the query based on the entity reference and based on the query comprises including at least one query term, from the query, in the natural language answer and including the particular text, of the entity reference, in the natural language answer.
11. The system of claim 8 , wherein the operations performed by the one or more processors in executing the instructions further comprise:
identifying an alternate entity reference from the content of the top ranked search result,
wherein the alternate entity reference includes alternate particular text that refers to an alternate particular entity that is also of the particular type of entity, and
wherein selecting the entity reference comprises selecting the entity reference over the alternate entity reference.
12. The system of claim 11 , wherein selecting the entity reference over the alternate entity reference comprises:
determining a first ranking signal for the entity reference and a second ranking signal for the alternate entity reference;
ranking the entity reference and the alternate entity reference based on the first ranking signal and the second ranking signal; and
selecting the entity reference based on the ranking.
13. The system of claim 8 , wherein the particular type of entity is the person type.
14. At least one non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
obtaining, for a query, search results for the query and a ranking of the search results,
wherein the query is a natural language query,
wherein the search results are generated based on the query, and
wherein the ranking of the search results is based on relevance of the search results to the query;
identifying, based on at least one term of the query, that the query is associated with a particular type of entity, of a plurality of candidate types of entities,
wherein the particular type of entity defines a particular categorization that encompasses multiple entities, and
wherein the particular type of entity is one of a person type, a location type, or a date type;
selecting one or more top ranked search results, as determined based on the ranking of the search results;
selecting an entity reference from content of at least one of the selected one or more of the top ranked search results,
wherein the entity reference includes particular text that refers to a particular entity, and
wherein selecting the entity reference is responsive to the entity reference being included in the content of the selected one or more of the top ranked search results and is responsive to the entity reference being of the particular type of entity identified based on the at least one term of the query; and
providing the entity reference for display along with, but visually distinguished from, at least the top ranked search results, wherein providing the entity reference for display at a computing device is responsive to a submission of the query from the computing device, and wherein providing the entity reference visually distinguished from at least the top ranked search results is responsive to selecting the entity reference.
15. The at least one non-transitory computer-readable medium of claim 14 , wherein providing the entity reference for display along with, but visually distinguished from, at least the top ranked search results comprises providing the entity reference positioned above the top ranked search results.
16. The at least one non-transitory computer-readable medium of claim 15 , wherein the entity reference is provided for display as part of a natural language answer to the query, and the operations further comprise:
generating the natural language answer to the query based on the entity reference and based on the query, wherein generating the natural language answer to the query based on the entity reference and based on the query comprises including at least one query term, from the query, in the natural language answer and including the particular text, of the entity reference, in the natural language answer.
17. The at least one non-transitory computer-readable medium of claim 14 , wherein the entity reference is provided for display as part of a natural language answer to the query, and the operations further comprise:
generating the natural language answer to the query based on the entity reference and based on the query, wherein generating the natural language answer to the query based on the entity reference and based on the query comprises including at least one query term, from the query, in the natural language answer and including the particular text, of the entity reference, in the natural language answer.